Tumor cells often appear in a deviant stage of differentiation, and dedifferentiation is a hallmark of malignancy. Alteration of gene expression by aberrant DNA methylation on promoter CpG islands and histone acetylation of chromatin are two mechanisms identified for epigenetic changes. Such types of epigenetic changes may account for changes of some gene expression in dedifferentiation; however, the causative mechanism of the global changes in dedifferentiation is not understood. Disabled-2 (Dab2) is a candidate tumor suppressor of ovarian cancer and its expression is lost in the majority of surface epithelial-derived tumors. Recent studies suggest that Dab2 functions in establishing epithelial polarity and surface positioning. The loss of Dab2 expression in tumor cells is not due to DNA methylation or histone acetylation; rather, Dab2 is regulated by the developmental transcription factors, GATA-4 and GATA-6, which are lost in ovarian tumor cells. We realize that the loss of GATA-4 and GATA-6 may underlie the dedifferentiation of ovarian surface epithelial cells in cancer, defined here as the loss of polarity and organization by a basement membrane. Thus, to test the hypothesis that developmental errors (the loss of GATA-4 and GATA-6 expression) lead to dedifferentiation and contribute to the neoplastic transformation of ovarian surface epithelial cells, we will determine the sequence of loss of expression of GATA factors, Dab2, laminin, and collagen IV in the pre-neoplastic morphological transformation of ovarian surface epithelium using pre-neoplastic lesions of contiguous epithelia linking morphological normal with neoplastic areas found in ovarian tumors (Aim 1). We will analyze the regulatory relationship between GATA factors and laminin, collagen IV, and Dab2 in cultured ovarian surface epithelial and tumor cells (Aim 2). Lastly, we will determine the consequence of losing GATA-6 and GATA-4 expression using ovarian-specific conditional knockout mice, and investigate the potential tumor-prone phenotypes of the animals (Aim 3). These studies will verify the hypothesis that the aberrant alteration of GATA-4 and GATA-6 underlies dedifferentiation of ovarian cancer and is a general early event in neoplastic transformation of ovarian surface epithelial cells. This study provides a novel mechanism for the contribution of developmental errors in the neoplastic transformation of ovarian surface epithelial cells. Such analysis may lead to new understanding of the tumor biology and etiology of ovarian cancer.